3
SSA's Proposed Design for the Disability Evaluation Study

In the previous sections we described briefly the background leading SSA to redesign the process by which eligibility for disability payments is determined, as well as the agency's approach toward the redesign plan. In this section we briefly discuss the general features of the Disability Evaluation Study (DES) as planned by SSA; identify the key survey design and sampling plan, data collection, and operational decisions made to date; and discuss some of their limitations as they relate to the efficiency of the sampling plan and accepted statistical principles and practices. We also lay out alternatives for SSA to consider and rule on prior to making final decisions about the design and operation of the DES in the scope of work for the survey RFP.

Information Goals of the DES

In response to a recommendation made by the Board of Trustees of the Federal Old-Age and Survivors Insurance and Disability Insurance Trust Funds in 1992 (DHHS, 1992), SSA has initiated a significant research effort, with both a short- and long-term focus, aimed at understanding the growth of the disability programs. DES is the cornerstone of SSA's long-term disability research plan. It is a complex multiyear national survey of the population 18-69 years of age.

The adequacy of any survey design depends on what questions the data collected are intended to answer. SSA has identified in the draft scope of work (SSA, 1996a, pp. 11-12) four major information goals for the survey:

  1. Estimate the total number and characteristics of persons in the United States with medical, vocational, and functional limitations or impairments, that is, persons with impairments severe enough to meet SSA's definition of disability; this group would represent the universe of potentially eligible nonbeneficiaries who could apply and meet the current criteria, but who are not now receiving benefits.
  2. Identify the number and characteristics of people who are not eligible under the current SSA definition of disability, but who could be included as a result of any changes in the disability decision process.


The National Academies | 500 Fifth St. N.W. | Washington, D.C. 20001
Copyright © National Academy of Sciences. All rights reserved.
Terms of Use and Privacy Statement



Below are the first 10 and last 10 pages of uncorrected machine-read text (when available) of this chapter, followed by the top 30 algorithmically extracted key phrases from the chapter as a whole.
Intended to provide our own search engines and external engines with highly rich, chapter-representative searchable text on the opening pages of each chapter. Because it is UNCORRECTED material, please consider the following text as a useful but insufficient proxy for the authoritative book pages.

Do not use for reproduction, copying, pasting, or reading; exclusively for search engines.

OCR for page 5
3 SSA's Proposed Design for the Disability Evaluation Study In the previous sections we described briefly the background leading SSA to redesign the process by which eligibility for disability payments is determined, as well as the agency's approach toward the redesign plan. In this section we briefly discuss the general features of the Disability Evaluation Study (DES) as planned by SSA; identify the key survey design and sampling plan, data collection, and operational decisions made to date; and discuss some of their limitations as they relate to the efficiency of the sampling plan and accepted statistical principles and practices. We also lay out alternatives for SSA to consider and rule on prior to making final decisions about the design and operation of the DES in the scope of work for the survey RFP. Information Goals of the DES In response to a recommendation made by the Board of Trustees of the Federal Old-Age and Survivors Insurance and Disability Insurance Trust Funds in 1992 (DHHS, 1992), SSA has initiated a significant research effort, with both a short- and long-term focus, aimed at understanding the growth of the disability programs. DES is the cornerstone of SSA's long-term disability research plan. It is a complex multiyear national survey of the population 18-69 years of age. The adequacy of any survey design depends on what questions the data collected are intended to answer. SSA has identified in the draft scope of work (SSA, 1996a, pp. 11-12) four major information goals for the survey: Estimate the total number and characteristics of persons in the United States with medical, vocational, and functional limitations or impairments, that is, persons with impairments severe enough to meet SSA's definition of disability; this group would represent the universe of potentially eligible nonbeneficiaries who could apply and meet the current criteria, but who are not now receiving benefits. Identify the number and characteristics of people who are not eligible under the current SSA definition of disability, but who could be included as a result of any changes in the disability decision process.

OCR for page 5
Identify the factors (i.e., accommodations, social support, and other factors) that permit persons with similar impairments who could qualify for benefits to continue working. Identify the variables needed to monitor and assess in a cost-effective manner future changes in the prevalence of disability. The committee has reviewed these goals in the light of alternatives to conducting a survey on a nationally representative sample. It judged that the informational needs of SSA regarding the size and characteristics of the population eligible for benefits, factors permitting them to work, and assessments of future changes in the prevalence of disability could not be fulfilled easily using other existing information sources. RECOMMENDATION 3-1. The committee strongly endorses the conduct by the Social Security Administration of a well-designed, carefully pretested and statistically sound Disability Evaluation Study. Large scale surveys like the DES are complex undertakings. They require careful planning prior to the data collection phase and refining during and after data collection. A primary principle of survey design is to determine in considerable detail what data are to be collected and how the data will be used so that decisions can be made on sample design, reliability, and collection procedures. Many federal agencies, when mounting such surveys, establish a committee of technical advisors to offer in the short-term technical input to ensure that management decisions include a careful consideration of possible alternatives. Reliance on the contractor alone to provide such input might not serve SSA well. The committee urges SSA to avail itself of such technical input in a structured manner. In making this recommendation the committee also urges SSA to place the DES in the context of the ongoing statistical needs of the agency. For example, SSA has stated that it hopes the DES will permit better forecasting of the changes in size of the beneficiary population. This implies ongoing measurement of the size of the pool, with updated instrumentation to reflect any changes in the SSA eligibility protocol. The DES as currently designed does not provide such a capability. Thus, SSA needs to articulate, preferably before all aspects of the DES have been fixed, the methods of providing ongoing monitoring of the size of the eligible pool of beneficiaries. The committee also believes that a detailed plan for the statistical analysis of the DES is required. This plan should include the development and validation of models that forecast the size of the disabled population and that synthesize the DES data with other data sources. General Features of the DES Design As originally conceived, the principal information goal of the DES was (a) to estimate the size of the population potentially eligible for disability benefits in order to assess the upper bounds for the growth of SSDI and SSI programs, and (b) to identify the factors that enable some persons with severe impairments to remain in the workforce.

OCR for page 5
A national probability sample survey to obtain this information is planned. In order to efficiently identify a relatively large group of potentially eligible nonbeneficiaries a decision was made to use a screening mechanism. These nonbeneficiaries are persons whose impairments are severe enough that they would likely be eligible for disability benefits if they applied. Other subgroups—current beneficiaries, people with lesser impairments (the ''borderline'' group), and nondisabled persons—are included in the survey to ensure full coverage as well as to provide the data needed to meet the DES objectives. Westat, Inc., under contract with SSA, developed a protocol for the conduct of the DES, setting out and weighing the relative merit of different options to address key aspects of the survey design. The protocol included an appraisal of the basic parameters of the overall size and scope of the survey (Westat, Inc., 1995b). DES in Three "Stages" While the DES was being developed, efforts to reengineer the process were underway and the disability decision process was being redesigned on a parallel but separate track. DES assumed an additional role of evaluating the proposed redesigned process and of serving as a source for testing functional assessment instruments and the decision process itself. The protocol proposed by Westat was modified to accommodate these additional roles of the DES. A major alteration, as conceived by SSA, is to conduct the survey in three distinct consecutive stages: a pilot test followed by two separate national data collection stages. Table 3-1 summarizes SSA's survey plan in three "stages" as described in their draft scope of work. As indicated in Table 3-1, SSA plans to conduct a comprehensive pilot test about 6 to 9 months after awarding the contract and completing the development work for the survey, and prior to launching stage 1. The pilot is meant to test operational methods and data collection instruments, response rates, respondent burden, and to determine what changes are required. The draft scope of work and subsequently reported timelines indicate a 3-month period for the pilot study, but the committee understands that final decisions on the scope and duration of the pilot study have not been made. A 3-month period would be sufficient for little more than a dress rehearsal of a close-to-final-form survey design. Given the present state of the DES protocol, this use of a pilot test stage seems premature. The committee believes it would be more realistic for SSA to use the entire time period set aside for the current stage 1 to develop and field test a DES protocol that would be fully implemented on a national basis in what SSA calls stage 2. SSA also indicates in the draft scope of work that the pilot test cases and data collection will be used as part of the stage 1 sample, if possible. Considerable changes in the final questionnaires and in operations are usually required as a result of pretests and pilot tests. The committee therefore thinks that including pretest cases in a major national survey like the. DES is not wise. As currently planned, both stage 1 and stage 2 of the survey are to be designed as complementary national samples so that nationally representative estimates can be generated from each individual sample as well as from both samples combined. SSA's design specifications for the two samples require that they be approximately equal in size and drawn from separate, nonoverlapping primary sampling units (PSU). SSA expects to combine the data from the two stages for most of the survey analysis. The stated purpose of the two separate data collection stages is to accommodate the reality that some of the components of the revised disability decision process that need to be tested in

OCR for page 5
TABLE 3-1 Current Survey Design Proposed by the Social Security Administration STAGES ACTIVITY Pilot Test A comprehensive pilot test of all operational methods and data collection instruments to assess what if any, protocol alterations are required. Stage 1 Half of Data Collection—First half of DES data collection (including telephone screening, interviews, clinical examinations, functional assessments, other tests, etc.) applied to a national half sample.   Final Changes in Redesign Method—made in conjunction with the pilot, stage 1 data collection, as well as with separate laboratory testing, this research would facilitate final resolution of issues tied to the redesign decision process, e.g., developing an acceptable functional assessment screening instrument, to experiment with the use of Department of Labor's O*NET for job requirements.   Supplemental Research—As needed and separate from DES data collection, further research would be undertaken to address other process and design issues. Stage 2 Half of Data Collection—Second half of DES data collection (including telephone screening, interviews, clinical examinations, functional assessments, other tests, etc.) applied to a complementary national half sample; part of the stage 2 survey instrumentation would be determined by the outcome of stage 1.   Analysis—of national and subnational profiles of disability using stage 1 and 2 data combined, assessment of the redesigned decision process.   SOURCE: SSA, 1996b. stage 1 and will not be completed until after the fieldwork for stage 1 is well under way. SSA, therefore, plans to collect data in stage 1 to facilitate early testing and refining some of the proposed functional assessment criteria for the redesigned disability decision process. In stage 2, a second nationally representative sample will employ the functional assessment instruments presumably improved from stage 1. SSA expects stage 2 to provide a direct field test of the redesigned disability decision process under development. The interview, medical examination, and functional assessment instruments used in stage 1 will also be administered to the stage 2 sample, as will any new or modified functional assessment instruments required for testing the redesigned disability decision process that were not available for inclusion in stage 1. In addition, stage 2 will complete the total sample size required to meet the goals of estimating the potential pool of eligible nonbeneficiaries, those of that group who continue to work, and the accommodations they receive. SSA states that most analytical uses of the DES will

OCR for page 5
combine the data from both stage 1 and stage 2 samples (SSA, 1996a, p. 16). The committee seriously doubts that pooling the data from the two samples is scientifically defensible. The dynamic state of the DES survey instruments is bound to make the stage 1 and stage 2 questionnaires substantially different, thus leading to differential context effects in a large number of survey questions (Schuman and Presser, 1981; Turner and Martin, 1984; Groves, 1989; Smith, 1991; Tanur, 1991). Because stage 1 and stage 2 will involve somewhat different methods, operations, and instruments, it will not be possible to combine the data as planned without introducing "stage effects" in DES estimates resulting from data inconsistencies. After changes are made in stage 2, the data that are collected from the two stages will not be comparable. Content and administration of the questionnaires and functional assessment instruments will differ in the two stages and some questions and tests (e.g., the final functional assessment instruments) will be administered in stage 2 but not in stage 1. The committee understands that stage 1 is meant to identify and test the measures of functional ability, and to inform the development of the final instruments for use in stage 2. However, a nationally representative sample is not needed for these purposes. Much of the work on definitions and procedures for data collection could be tested with small samples using different approaches including the use of laboratory research. Increasingly survey operations, especially for large scale surveys, are being preceded by small scale, carefully structured field testing coupled with laboratory research, so that definitions and concepts critical to the results can be better understood. Geographically, socially, and educationally dispersed nonprobability samples, laboratory subjects, or both are adequate to develop and pretest the functional assessment instruments. Indeed, several existing sources can be used for such testing for the DES.3 RECOMMENDATION 3-2. The committee recommends that the current stage 1 and pilot study be merged, expanded, and extended into a research, development, and testing phase of the survey with application to samples of the type that are more traditionally used in methods testing. Only When the development and refinement of the functional assessment instruments, survey operations, and other issues are tested and resolved should a national sample survey be launched using a single protocol. Recommendations 3-2 and 3-3 are part of a broad revision of the survey design for the DES. Table 3-2 presents a summary of the committee's recommended survey design. A significant advantage of these recommended changes in survey design for the DES is that they will ensure a more statistically sound design and accompanying estimates, with no significant delay in the award of the contract. The research, development, and testing phase will be part of the survey contract. In addition, they are likely to result ultimately in appreciable cost savings (since one rather than two national surveys would be conducted (Table 3-3). 3   Some examples of these sources are state vocational rehabilitation agencies, labor unions, large companies, federal government agencies, and national associations that advocate for those with impairments such as Independent Living Centers, the Association of Retarded Citizens, the Learning Disability Association, the Manic-Depressive Society, the National Amputee Coalition, People First, and the Paralyzed Veterans of America.

OCR for page 5
TABLE 3-2 The Committee's Recommended Survey Design RESEARCH, DEVELOPMENT, AND TESTING NATIONAL PROBABILITY SURVEY Pretests—Field testing of instruments, procedures, modes of examinations, alternative approaches to screener and response rates, and functional assessment instruments on small subnational, nonprobability samples. Data Collection—Data collection (including screening, interview, examinations, functional testing and other tests, etc.) applied to a national probability sample of size equaling or exceeding the size of the samples for stages 1 and 2 combined. Comprehensive Pilot Test—Testing of the survey instruments, examination, functional assessment criteria, and survey procedures, possibly followed by a dress rehearsal of the survey. Analysis—National and subnational profile of disability using DES data; assessment of redesigned decision process. Final Changes in Redesign Method—Done as a result of the pretesting and pilot testing, as well as relevant laboratory testing. This research would facilitate, prior to the conduct of the main survey, final resolution of issues tied to the redesigned decision process, e.g., developing an acceptable functional assessment instrument and experimenting with the use of Department of Labor's O*NET for job requirements.   Analysis of Existing Data—Using existing national data sources ( e.g., NHANES, NHIS, SIPP, etc.), complete analyses to inform SSA and the contractor about sample issues and that at least partially address some of the DES information goals.  

OCR for page 5
TABLE 3-3 Advantages of the Recommended Survey Design   Retain current timeline—SSA can still award the contract following the recently revised timeline. More time for preparation and resolution of issues for DES data collection—More needed time would be available to resolve remaining decision process and survey design issues, including development of functional assessment instruments, in advance of beginning the national DES data collection. Larger sample size for national estimates—National estimates would be derived from the full national sample drawn from a larger number of primary sampling units, as opposed to the half samples containing data based on different questions and testing instruments. Uniform data sets—One uniformly gathered set of data would be available, rather than two half samples with less than fully comparable data. Stronger design/redesigned comparisons—Data would be available for assessing the impact of design/redesigned decision processes applied to the same set of respondents. Lower data collection cost—Although the presurvey period will be longer then currently planned and the cost for this activity may be higher, ultimately cost savings would likely arise from having one rather than two data collection startups and tests done on non-national samples. Sample Design The sample design for the DES (both stage 1 and stage 2 of the survey) as planned by SSA is driven by the following four objectives (SSA, 1996a, p. 20): produce precise estimates of the various subgroups of the working-age population with disabilities and who among them are disabled enough to be likely to be eligible for disability benefits if they applied; yield a sample of nonbeneficiaries with disabilities sufficient to permit analyses by a number of subclasses; yield a sample of the "borderline" group of persons with disabilities sufficient to permit estimates of the number and characteristics of persons who might become eligible, or cease to be eligible, if the current SSA disability decision criteria are altered, and yield a sample of nondisabled persons sufficient to permit comparisons with the population with disabilities on measures of physical and functional performance and medical conditions in the population.

OCR for page 5
TABLE 3-4 Summary of Sample Design Proposed by the Social Security Administration SAMPLING STEP* SAMPLING UNITS STRATIFICATION TYPE OF SELECTION METHOD APPROXIMATE SAMPLE SIZE 1 Contractor-specified primary sampling units (PSUs) Unspecified by SSA (typically by region, racial/ethnic mix, and degree of urbanization) Unspecified by SSA (typically with probability proportional to population size or number of households) 100-110 PSUs mentioned by Westat—half of which will be used for each data collection stage 2 (Phase 2) Households (from a sample of residential phone numbers; beneficiary sample from SSA records) Unspecified by SSA (beneficiary status, if SSA records are used as a frame source) RDD Sampling (SSA's choice, although face-to-face screening was considered) 171,000 persons 18-69 years of age (in 100,000 screened households) 3 (Phase 2) Person Yes Unspecified 5,500   (every person in household up to 3 persons) By disability groupings: (Stratified random sample, with variable sampling rates set to achieve targeted sample sizes for the four disability groups) By disability groupings (in Col. 2):       Potentially eligible nonbeneficiaries Borderline nonbeneficiaries (persons with lesser impairments) Nondisabled persons Current beneficiaries     Potentially eligible nonbeneficiaries (3,000): who are working persons (700) who are younger (450-600) who are 62-69 years old unspecified) with mental, emotional or behavioral conditions (750-900) from minority groups (500 blacks) Borderline nonbeneficiaries (persons with lesser impairments) (1,500) Nondisabled persons (500) Current beneficiaries (500) *These levels of sampling, called "stages" in sampling statistics, are referred to here as "steps" to minimize confusion with SSA's reference to "stages'' when referring to "rounds'' of DES data collection. SOURCE: SSA, 1996b.

OCR for page 5
The sample sizes proposed by SSA appear to be driven primarily by the first objective and by cost considerations. With those two factors in mind, SSA has set a goal of identifying a sample of 3,000 nonbeneficiaries with severe disabilities (the likely eligible group), out of a total sample of 5,500 persons. Table 3-4 summarizes the sample design as planned by SSA and the size of the subgroups within the total sample. Severe impairments are relatively rare in the general population. In fact, the severity and prevalence of a disabling condition are inversely related; the higher the prevalence of a condition, the lower the severity, and vice versa (LaPlante, 1991). Because SSA's eligibility criteria tend to filter out people with less severe disabilities, SSA is faced with many low-prevalence disabling conditions all of which cannot be screened adequately into the sample. The exceptions may be mental conditions and low back conditions. SSA is cognizant of this situation, and therefore it has built into its sampling plan provision for oversampling persons with severe disabilities. Accordingly, the sample will contain: a "core" group of nonbeneficiaries with severe disabilities (3,000), persons with significant but lesser impairments, the "borderline" cases (1,500), nondisabled persons (500), and current SSDI and/or SSI disability beneficiaries, who will be included primarily for the purpose of benchmarking the distinctive characteristics of the core group (500).4 SSA assumes that the core group sample of 3,000 will be sufficient to analyze several subgroups of particular policy interest. These subgroups will include potentially eligible nonbeneficiaries who are working; younger nonbeneficiaries with disabilities; nonbeneficiaries aged 62-69 years; nonbeneficiaries with mental, emotional, or behavioral conditions; and nonbeneficiaries with disabilities from minority groups. The committee has several questions and concerns about the adequacy of the size of the total combined sample, of stage 1 and stage 2 samples, and of the allocations among the four subgroups. How did SSA arrive at this particular disproportionate sample design? The allocation clearly favors the 3,000 nonbeneficiaries with severe disabilities. What is the basis for choosing the sample sizes for the four groups? What precision targets were used to arrive at the sample sizes? What response rates are to be reached for each component of the DES and for the overall survey? It appears that the sample sizes discussed in stages 1 and 2, both separately and combined, will lack the condition specificity that SSA would require for estimation and analytical purposes. The cells will be much too small, especially if SSA stratifies on more than one disabling condition and/or demographic or socioeconomic characteristics such as age, gender, minority status, or working nonbeneficiaries with specific disabling conditions. Similarly, the proposed sample size for the borderline group of persons with less severe disabilities may not be sufficient in its analytical strength for assessing how alternative decision processes or definitions of disability would affect outcomes. The differences in outcomes resulting from changing the decision process is likely to be minimal if any for persons with severe disabilities, but some real differences could show up among borderline cases under alternate 4   According to the draft scope of work, this sample of current beneficiaries may be drawn from screened cases, from SSA administrative records, or both.

OCR for page 5
decision processes. Clearly, SSA needs to give more thought to issues of sample selection, size, and allocation. RECOMMENDATION 3-3. The committee recommends that the national survey should be conducted with one sample large enough to estimate the sizes of the populations at risk with acceptable levels of statistical precision. This recommendation, along with Recommendation 3-2, reflects the committee's concern about the plausibility of merging data from stages 1 and 2 of the DES. Also, they reflect the inadequacy of a 1-year period after award of the contract for planning, developing, and testing the survey methods that will be used for the national DES. The impact of these two recommendations on the timeline of the DES and on SSA's research on the proposed changes in the disability decision process are shown in Figure 3-1. There one can see that the committee's recommendation to gather DES data as part of a single national survey, rather than as part of two potentially incompatible half samples, implies a period of at least 24 months instead of 12 months for the research, development, and testing activities. These activities will be a key to the success of the DES. Analysis of Data from Existing National Surveys Over the past many years, several national surveys have included a significant number of questions through special modules or supplements about chronic disease, disability, employment status, and receipt of disability benefits. Findings from the surveys provide some information about the relationships among these factors, but the usefulness of the information for purposes of SSA's disability decision process redesign is limited by several problems, including the small number of working-age subjects with severe disabilities, the lack of sufficiently detailed questions about their impairments, and the absence of a medical examination and tests of functional capacity. However, data from these surveys could provide some idea of the prevalence of disability in this country and the proportion and characteristics of those people who are working despite their disabilities. Such data could be useful to guide SSA in questionnaire development and in designing the sampling plan for the DES. Some examples of major surveys are the National Health Interview Survey (NHIS) and the National Health and Nutrition Examination Survey (NHANES) both conducted by the National Center for Health Statistics (NCHS), and the Survey of Income and Program Participation (SIPP) conducted by the Bureau of the Census. The 1994-1996 Disability Supplement of the NHIS was conducted by the NCHS in two phases with a nationally representative sample of people 18-64 years of age. It addresses many of the issues of interest to DES. The Phase I and II questionnaires include detailed questions about chronic disease, disability, employment status, and receipt of disability benefits. The sample for Phase II of the survey includes more than 10,000 working-age persons with disabilities based on an extensive interview in Phase I.5 5   The 1994 Phase I data collection was completed in December 1995, and data from the 1994 Phase I interviews are currently available for analysis. The 1994 Phase II data collection is almost complete. The 1994 Phase II data on adults will be available by July 1997. The 1995 Phase I data collection will be completed by June 1997, and Phase II will be done by the end of summer.

OCR for page 5
Figure 3-1 Alternative timeline for the Disability Evaluation Study. (The timeline for the decision process research will not be affected.)

OCR for page 5
The committee's preliminary review of the questionnaires for Phases I and II of the NHIS Disability Supplement yielded many questions that seem relevant for the development of the DES. Table 3-5 shows the topics covered. Within each topic, the questionnaires include many detailed questions. The committee recognizes that the NHIS Disability Supplement does not include performance tests, medical examinations, or subjects' medical records. Thus, the survey findings cannot be used to simulate SSA's current disability decision process—one of SSA's objectives for the DES. On the other hand, analysis of the survey findings would provide valuable information from a nationally representative sample on (a) the proportion of working-age people with certain diseases, conditions, and functional limitations who do and do not work, (b) the proportion of those who do not work who receive disability benefits, and (c) respondents' perceptions of factors that affect their ability to work. This information would be useful in designing the screening instrument for DES, particularly in developing questions that would help to identify individuals in two of SSA's target groups: nonbeneficiaries who might be expected to meet the SSDI/SSI criteria if they were not working, and people in the borderline group, who might be expected to nearly meet the criteria.6 A topical supplement containing an extensive set of questions about disability was used as part of the sixth wave of the 1990 panel and the third wave of the 1991 panel of the SIPP. The total sample size for this study was approximately 30,000 interviewed households (McNeil, 1993). Similarly, NHANES-III, conducted from 1988 to 1994 on a sample of 40,000 persons aged 2 months and older, included a relatively small number of working-age subjects with severe disabilities; it did, however, include medical examinations and some performance tests (NCHS, 1994). The committee believes that experience gained from this survey could answer important questions about response rates, refusals, and procedures for conducting medical examinations and performance tests on subjects with severe disabilities in a survey context, and about topics such as self-versus proxy reports. Moreover, SSA could learn from the experience of the NHANES survey, in particular, about conducting medical examinations and performance tests on people with severe disabilities. For example, people with severe disabilities may not be willing to leave home for these procedures, and may not want to undergo a lengthy medical examination. Others may not be able to provide accurate information about their medical records. Proxy respondents and home examinations may be a necessity for persons with severe disabilities. RECOMMENDATION 3-4. The committee recommends that the Social Security Administration use relevant data from the National Health Interview Survey Disability Supplement, National Health and Nutrition Examination Survey, Survey of Income and Program Participation, and other relevant surveys to assist in developing the sample design, survey operation, and questionnaire content for the Disability Evaluation Study. 6   The 1994-1996 Disability Survey also includes a nationally representative sample of elderly persons. The Phase II interview for the elderly differs in some ways from the Phase II interview for the nonelderly. It is likely, however, that information about subjects age 65-69 could be used to answer some of SSA's questions about people in this age group.

OCR for page 5
TABLE 3-5 Topics and Variables from Phases I and II of the 1994-1996 Disability Survey, Population 18-64 Years of Age PHASE I: TOPICS PHASE II: TOPICS AND VARIABLES Sensory impairments (problems with seeing, hearing, smelling, tasting) Repeated Communication problems   Learning problems   Specified diseases and conditions Specified diseases and conditions ADL impairments (need for hands-on help or supervision, extent of difficulty Repeated with many variables, including difficulty, pain, and speed of performance IADL impairments (need for hands-on help or supervision, extent of difficulty Repeated with many variables, including difficulty, pain, and speed of performance Other functions and activities (lifting, walking, standing, bending, reaching, using fingers to grasp or handle objects) Repeated with many variables, including the difficulty with the function or activity Mental and emotional problems (impact on ability to find or keep a job and kind or amount of work the person can do)   Participation in sheltered workshop, work training, and supported employment Repeated with many variables Use of physical or occupational therapy Repeated with many additional services Use of vocational rehabilitation services Repeated with many variables Use of assistive devices Repeated with many variables Income and assets   Receipt of disability payments   Employment status Employment status or work history (ever worked, type of work, volunteer work)   Need for accommodations in order to work (e.g., ramps, parking, special workstation)   Need for other assistance to work (e.g., interpreter, job coach, personal attendant)   Reasons for not working or looking for work (e.g., fear of loss of disability benefits, or health insurance, lack of transportation)   Type of living situation (e.g., apartment, group home, center for independent living); adaptations or modifications to home, services provided by living situation   Transportation (ability to drive, use of regular public transportation, special transportation, mobility training)   Social activities (e.g., visit friends or relatives, go out, talk on the telephone)   SOURCE: NCHS, 1995.